The Department of Tumor Surgery, Lanzhou University Second Hospital, Lanzhou, Gansu, China.
The Second Clinical Medical College, Lanzhou University, Lanzhou, Gansu, China.
Cell Biol Int. 2023 Sep;47(9):1650-1664. doi: 10.1002/cbin.12053. Epub 2023 Jun 27.
Reliable prognostic signatures that can reflect the intrinsic characteristics of gastric cancer (GC) are still rare. Here, we developed an adenosine-based prognostic signature and explored its association with the tumour immune in GC patients, aiming at confirming the prognostic value of adenosine-related genes and guiding the GC risk stratification and immunotherapeutic response prediction. We collected adenosine pathway-related genes from STRING websites and manual searching. We enrolled the The Cancer Genome Atlas cohort and four gene expression omnibus cohorts of GC for generating and validating the adenosine pathway-based signature using the Cox regression method. Gene expression in the signature was verified using polymerase chain reaction. We also performed gene set enrichment analysis, immune infiltration assessment and immunotherapy response prediction based on this signature. Our study resulted in a six-gene adenosine signature (GNAS, CXCR4, PPP1R1B, ADCY6, NT5E and NOS3) for risk stratification of GC prognosis, with the highest area under the receiver operating characteristic curve up to 0.767 for predicting 10-year overall survival (OS). In the training cohort, patients with signature-defined high risk had significantly poorer OS than those with low risk (p < .001). Multivariate analysis identified the signature as an independent prognostic factor (hazard ratio 2.863, 95% confidence interval [1.871-4.381], p < .001). These findings were confirmed in four independent cohorts. Expression detection showed that all signature genes were upregulated in both GC tissues and cell lines. Further analysis revealed that the signature-defined high-risk patients were characterised by immunosuppressive states and associated with a poor immunotherapy response. In conclusion, the adenosine pathway-based signature represents a promising risk stratification tool for GC in guiding individualised prognostication and immunotherapy.
可靠的预后标志物，能够反映胃癌（GC）的内在特征仍然很少。在这里，我们开发了一种基于腺苷的预后标志物，并探索了其与 GC 患者肿瘤免疫的关联，旨在确认与腺苷相关基因的预后价值，并指导 GC 的风险分层和免疫治疗反应预测。我们从 STRING 网站和手动搜索中收集了腺苷通路相关基因。我们纳入了 The Cancer Genome Atlas 队列和四个 GC 的基因表达综合队列，使用 Cox 回归方法生成和验证基于腺苷通路的标志物。使用聚合酶链反应验证了该标志物中的基因表达。我们还基于该标志物进行了基因集富集分析、免疫浸润评估和免疫治疗反应预测。我们的研究产生了一个用于 GC 预后风险分层的六基因腺苷标志物（GNAS、CXCR4、PPP1R1B、ADCY6、NT5E 和 NOS3），预测 10 年总生存率（OS）的受试者工作特征曲线下面积高达 0.767。在训练队列中，具有标志物定义的高风险的患者的 OS 明显差于低风险的患者（p<0.001）。多变量分析确定标志物是一个独立的预后因素（危险比 2.863，95%置信区间[1.871-4.381]，p<0.001）。这些发现在四个独立的队列中得到了证实。表达检测表明，所有标志物基因在 GC 组织和细胞系中均上调。进一步分析表明，该标志物定义的高风险患者具有免疫抑制状态，并与较差的免疫治疗反应相关。总之，基于腺苷通路的标志物代表了 GC 个体化预后和免疫治疗指导下有前途的风险分层工具。
